A Multiplicity of Management Systems: Performance management work,
recognition and accommodation in a multi-national corporation
David J Cooper
(School of Business, University of Alberta, Edmonton,
Alberta, Canada)
email: david.cooper@ualberta.ca
Mahmoud Ezzamel
(Cardiff Business School, University of Cardiff, Cardiff, UK)
email: ezzamel@cf.ac.uk
Keith Robson
(Cardiff Business School, University of Cardiff, Cardiff, UK)
email: RobsonK@cf.ac.uk
November 2011
The Social Science and Humanities Research Council and the Certified General Accountants of
Alberta support David Cooper’s work. We would like to thank Barbara Townley for her involvement
in the research which informs some of the arguments in this paper, and the managers of the firm who
participated in this study.
A Multiplicity of Management Systems: Performance measurement work,
recognition and accommodation in a multi-national corporation
“The choice of performance measures is one of the most critical challenges facing
corporations” (Ittner & Larcker, 1998: 205).
Organizational structures and management systems have been growing objects of
examination and prescription. Studies of Du Pont and General Motors, for example,
reinforced the attention given to the multi-divisional form (M-form) and the requisite
decentralized management systems (Chandler, 1962). The historical analysis of the
emergence of the M-form has been followed by close examination of the rise of the
multi-divisional form as the dominant organizational structure in the twentieth
century (Fligstein, 1991). M-form ‘logic’ highlighted the dual requirements of
decentralization or responsibilization, and integrative management systems that could
both guide/support divisional objectives and measure/compare their attainment. In
short there is widespread agreement that management systems in large M-form
organizations have a central role in mediating the relationship between a parent or
HQ, and divisional sub-units without which the logic of divisionalization cannot be
said to function (Solomon, 1985; Ezzamel, 1992; Ezzamel & Hart, 1987). Yet it is
also apparent that the logic of divisionalization is partly premised upon a notion of
autonomy from central control (Solomons, 1985; Ezzamel, 1992). Within broad
parameters of a decentralizing philosophy, divisionalization has tended to imply a
loose coupling of operative controls between centre and division/periphery (Vancil,
1979).
Our research is concerned with the management and control systems in a large
European Corporation and in particular the workings of the performance
measurement systems used within and between a divisional unit and HeadQuarters
(henceforth HQ). Ittner and Larcker called for research that can provide “a rich
description of emerging performance measurement practices,” (1998: 204) and that is
certainly one aim of our paper. More specifically, we explore the emergence and coexistence of multiple performance management systems side-by-side within the same
divisionalized organization or other part of an organization, such as a plant or a sales
unit – a phenomenon that has been noted by only a few recent studies (Ezzamel et al.,
2004). Yet such studies have frequently stopped short of addressing the phenomenon,
which is curious given potentially stimulating contrasts between the traditional views
2
of management information and performance measurement models as holistic systems
for rational planning, decision-making and control, and the apparent diversity that can
exist within organizations.
Since the development of contingency approaches to management accounting and
control (Cooper, 1981; Otley, 1980) there has been recognition that a control system
comprises many elements and each of these elements may relate to different
contingencies. These elements have been described as a control package, which has
been defined as follows: “As a general conception, a management control systems
(MCS) package is a collection or set of controls and control systems. The individual
control systems may be more traditional accounting controls such as budgets and
financial measures, or administrative controls, for example organization structure and
governance systems, along with more socially based controls such as values and
culture. Organisations may have numerous controls present, and they all may be used
to some extent to align individual’s activities with organisational goals (Malmi and
Brown, 2008: 287). Yet, despite the recognition of control packages, little empirical
work has been conducted to explore their interaction and use in practice, and what
research has been done (as for example in the special issue of Management
Accounting Research in 2008) tends to be focus either on specific types of
organizations or particular management practices.
The existence of a multiplicity of management systems thus raises a number of
intriguing issues that deserve the attention of researchers. First, it would be instructive
to know more of the processes that explain why multiple management systems have
diffused and co-exist in the same organization. Cooper et al (1996) refer to the idea of
sedimentation to explain the existence and persistence of different control approaches
in an organization. Second, this co-existence of multiple management systems is
suggestive of a variety of possible relationships that may arise from interaction
between these systems. As Malmi and Granlund (2009) suggest, we need research
that examines in what ways and how do multiple parts of a control package overlap,
replicate, collaborate, complement, or compete with each other. How do organizations
and their sub-units rationalize multiple systems? Other than a few notable exceptions
(Oakes et al, 1998; Ezzamel et al, 2004; Ezzamel and Burns, 2005; Kennedy and
Widener, 2008), we know little about how systems of management interact with one
3
another, and with how they may get played out in specific organizational settings.
Third, there is the issue of the possible impact of such multiplicity upon
organizational actors; e.g. confusion, uncertainty, clarity, autonomy, etc. What is at
stake for managers in opting for particular Performance Management Systems and in
ignoring others? In short, can we understand the logic or logics of a multiplicity of
management systems in divisions of the same organizations?
Mainstream management control systems theory of course highlights the centrality of
performance management to the rational planning and control of the enterprise
(Merchant & Van der Stede, 2011). Nevertheless, the economics underlying this
prescription suggests that optimal control should be achieved when the benefits of
avoiding control loss outweigh the costs of implementing controls (ibid., p. 10).
Where there are multiple and overlapping controls this might suggest issues of
(in)efficiency or significant problems of organizational control. In contrast, neoinstitutional theories suggest that such rational systems of control may have more to
do with organizational legitimation than operative consequences (Meyer & Rowan,
1977). In so doing much of the emphasis of neo-institutional work has been upon the
prevalence of organizational forms and practices within organizational fields: how
models of the appropriate actions or structures come to diffuse and be taken for
granted (Powell & DiMaggio, 1991; Fligstein, 1991). In more recent times there has
emerged a closer concern with variations in practices and forms, partly as a result of
interest in logics and actors’ rationales (Marquis and Lounsbury, 2007; Lounsbury,
2008), but also in regard to the channels of diffusion of models and practices, and
how some organizations pick up particular models while organizations in the same
field (and even organizational units in the same organization) appear to ignore them
(Strang & Meyer, 1993).
In this study we investigate empirically how an organization may be subject to many
of these competing pressures for operative control and institutional conformity but
with particular regard to the tensions not only between an organization and its
environment or field, but also internally: how these pressures and tensions may result
in complex and variegated management control systems, all seemingly functioning
but with different rationales and outcomes for corporate HQ and divisions. Moreover
we explore how these variations not only arise, but also play out in complex tensions
4
between the central HQ and the peripheral division (Ghoshal & Bartlett, 1988). We
show how a multiplicity of management technologies or elements not only emerges
but how it is to be ‘managed’ inside the corporation by the various users. Central to
our analysis are concepts of ‘work’, accommodation and recognition…
In the next section we consider the role of performance management systems in
organizations and outline the questions guiding our research. This is succeeded by a
discussion of the research methods and the research site. We detail the management
systems that we found in use within our research subject/corporation and then outline
in the central case section our analysis of the multiplicity of management systems that
we observed. This is followed by a discussion section in which we develop the key
concepts that we take from the case and which we then develop in relation to extant
theories relevant to the role of management system in organizations. In the concluding
section we suggest research propositions for further research and the limitations of
our analysis.
Performance Measurement Models and the Organizational Performance
Modern management texts and research are replete with advice on the importance of
performance measurement models and management to the attainment of
organizational performance (Merchant & Van der Stede, 2011). As Malina and Selto
note, consultant and academic normative prescriptions predict that systematic
performance management models and measures will achieve superior performance
and strategic direction (ibid., 2004: 442; Huff and Jenkins, 2003). From the early
example of the ROI concept in Du Pont (Johnson & Kaplan, 1987) to modern models
such as the BSC (Kaplan and Norton, 1992), EVA (Stern-Stewart, 2003) and Kanji’s
business scorecard (Kanji & Moura e Sa, 2002), models of business inputs, and
outputs, and measures of operational, strategic, financial and non-financial
performance have been tied together into systems of management control for
organizations to follow. Part of this trajectory has been an increased concern that
‘traditional accounting measures’ should be supplemented with non-financial
measures (Johnson & Kaplan, 1987; Hope & Fraser, 2003).
The benefits of performance measurement models are said to include strategic
improvement (by development of the firm’ resources), organizational learning (testing
5
predictions and obtaining feedback on the relations between inputs and outputs) and
guide/incentivize employee behaviours, but only insofar as the attributes of the
measures fit normative notions of decision-making/informativeness, completeness,
objectivity, accuracy, etc (Milgrom & Roberts, 1995; Ittner & Larcker, 1998: Ittner et
al. 2002). Moreover, the benefits of such models are only said to accrue to the extent
that their benefits outweigh the costs (Simons, 1987, 2002; Merchant and Van der
Stede, 2011). Within the general commitment to the ideal of performance
measurement models, however, studies have posited that the design of such models
may empirically or should normatively be affected by firm strategy (Porter, 1980,
1985; Chenhall, 2003; Hemmer, 1996; Baiman et al., 2010).
Within the broadly economics based literature on management control we note
several key assumptions about the role and consequence of performance measurement
models. Performance measurement models are often viewed from the perspective of
the firm’s owners rather than decision usefulness for managers/agents (Feltham and
Xie, 1994). Models may be chosen or designed in accordance with specific strategic
or other contingent conditions (Chenhall, 2003), but most studies assume that
performance measurement models operate within the aims or strategic directions of
the whole firm. As such performance measurement models are studied quite often as a
unitary choice taken by senior management. The existence of multiple performance
measurement systems is barely recognized or, more likely, assumed to be sub-optimal
given the requirement for holistic, complementary and integrated management
systems (Malina & Selto, 2004).
Yet perhaps most notably, the empirical evidence that supports the normative claims
of the benefits of economic value or multiple measure (financial/non-financial)
management and performance measurement models is not robust (Ittner & Larcker,
1998). Many studies rely on self-reporting in order to validate claims of effectiveness
(e.g., Rucci, et al., 1998; cf. Ittner et al., 1997; Ittner & Larcker, 1998). And failures
of particular performance measurement systems is typically explained away by
reference to general appeals to problems of “implementation” (e.g., Stern et al, 1995;
Dambrin & Robson, 2011). This is despite the recognition that the empirical evidence
supporting the success of such models is insubstantial.
6
Part of our point here is that in modern management theory there is a strong
commitment to the idea that integrated performance management systems constitute
almost essential components of ‘best practice’ for optimal organizational
performance. Moreover, it is a theory produced, supported and perpetuated by an
economy of premier business schools, management consultants, business media and
their corporate clients (Kaplan & Norton, 1992; Stern-Stewart, 2003). For example, as
Ittner and Larcker note:
“Consulting firms battle over the superiority of their economic value
measures, charging that competitors’ measures have flaws that compromise
their predictive ability. (Myers 1996; The Economist, 1996).” (1998: 213-4)
Moving away from this kind of rational economic prescription, other social scientists
have paid greater attention to the practices and rationales that organizational actors
bring to bear within corporations. Thus, neo-institutionalist theorizing suggests that
the administrative character of corporations is influenced by the dominant logics in a
firm’s environment. The focus upon organizational measurement and decisionmaking can be said to reflect a corporate logic (Friedland & Alford, 1991; Thornton
and Ocasio, 1998), though perhaps it is also embedded now in broader societal logics
that structure organizational processes and structures (Meyer, 2009). In this reading
the emphasis that corporations place upon performance measurement systems has
possibly more to do with creating the semblance of rational and optimal organization.
Studies of, for example, budgeting have gone some way to affirming that
management control systems can have symbolic roles for corporations (Covaleski &
Dirsmith, 1988). The prevalence of many competing claims for economic value and
non-traditional measurement systems also suggests of the need for studies to consider
the fad and fashions dimensions of popular management innovations (Abrahamson,
1991, 1996; Abrahamson & Eisenman, 2008; Ezzamel & Robson, 2011).
Insights into the motives that companies and their divisions have in taking up
particular systems of management control might also be gained by exploring the
channels through which such ‘innovations’ disseminate and the social movements that
support them (Strang and Meyer, 1993). How performance management systems are
introduced into firms and by whom are some of the clues to understanding how in our
case study we have a divisional using, in various ways, four overlapping performance
management systems. As Strang and Meyer (1993) note, the diffusion of new
7
innovations is heavily related to the type of adopters and their theorization in
organizations, professional associations, etc. Our study explores the influence of
different organizational groupings on the choice of which performance management
is important and why.
As we detail in the next section, our case study concerns the usage of multiple
performance management system inside on division of a large multi-national. Our
study focuses on the nature and characteristics of four performance measurement
systems, the purposes to which each was put within the division and in the relations
between the division and company HQ. In broad terms our study attempts to explore
this multiplicity of management systems and the various roles that each system
assumed within the corporation. Questions as to how the firm managed to develop
four systems and the relations between each system were explored so as to explore the
various practices in which each system was embedded.
In the next section we set out the research method, and explain something of the
research site and the situation within which we explored the roles of multiple
management systems.
Research Method
This paper is based on a field study in two units of a major European multinational.
Our initial focus was on the spread of the Balanced Scorecard (BSC) and performance
management across units within the firm. One key purpose of the wider study was to
analyze how management systems such as the BSC are impacted by national and
professional traditions. In the course of this research we were struck by the variety of
measures within these units and their mutual interaction. Although our study looked at
divisions in three different national settings, for the purposes of this paper it was
sufficient to focus on one division in one national setting, albeit that the corporate
headquarters are located in another national jurisdiction.
We adopt a qualitative research methodology (Denzin and Lincoln, 1994), but one
where theory is a crucial orientating framework that helps to guide both data
collection and analysis (Silverman, 1985). Accordingly we use the organizing frame
8
of decision making, institutionalization and diffusion to make sense of, and structure,
our research data. For the construction of this paper 46 interviews with relevant
employees from the divisional units at varying organizational levels as well as with
the international headquarters were directly relevant, although during the project we
conducted over 90 interviews in total. The precise content of the interviews varied
depending on the experience and background of the subject, but interviews typically
lasted about 90 minutes, were tape recorded and transcribed. MORE ON THIS
Our case study also examined numerous internal documents and generic descriptions
of the various performance measurement systems found within the firm. For reasons
of confidentiality, the firm and its internal documentation must remain anonymous.
However, some of the measurement systems used in our multinational are
unsurprisingly somewhat generic (such as BSC), and hence we refer to them by name.
Systems developed internally by the company are given pseudonyms.
The Research Company
In order to contextualize the emergence of multiple management systems in our case
study, we provide a brief account of its history. Gigantic (a pseudonym) is a major
multinational corporation with headquarters in a European country, half a million
employees and a turnover of over £60 billion. The firm is a conglomerate with
interests in energy, and various industries, organized upon divisional lines. Gigantic
is organized around several major divisions, including the Electronica Division (a
pseudonym) that has around 70 factories distributed globally. We focus on two linked
units within the Electronica Division in the UK. The unit within Electronica, which
we call Automata (a pseudonym) manufactures standardized, low complexity
products, employs about 400 people and has a turnover of approximately £50M. The
other unit Mercado makes and sells a wide range of electronics products (including,
but not limited to the items produced by Automata), employs over 600 people, and
has an annual sales volume of about £150M. Our material focuses more closely upon
the manufacturing unit, though the sales unit was in close proximity to Automata and
shared the same characteristics in terms of the multiple management systems.
Interviews with managers at Mercado proved helpful in validating the rationales that
we noted for the ambiguous uses on different performance measurement systems. The
fortunes of Mercado as a sales unit was essentially intertwined with the success of
9
Automata and hence the later unit’s history had more significance in terms of the
preferences for different measurement systems that we noted in the two units.
In the next section we move to the case of Automata and their usage of performance
measurement systems. We start by first outlining, briefly each of the four
management systems that we encountered, before considering their uses, roles and
inter-relations within Automata, Mercado and Gigantic.
A Multiplicity of Management Systems: the control systems of Automata
Prelude: the four performance measurement systems
The accounts from our interviews and internal documentation that we examined
pointed to the co-existence of four management systems at Automata.
1.
2.
3.
4.
European Foundation for Quality Management (EFQM);
Balanced Scorecard (BSC);
Process Improvement (PI);
Time.
The first two listed are associated with specific performance management systems
that are ‘external’ in the sense that the underlying model was developed outside of
Gigantic and are ‘proprietary’. The other two were developed in-house by Gigantic.
We devote a little more attention to explaining the internal models as the external
ones are better-known generally and otherwise visible.
European Foundations for Quality Management (EFQM): EFQM derives from a
model generated by a consortium of European companies that was founded in 1988.
The EFQM is a global non-for-profit membership foundation based in Brussels,
Belgium. With more than 500 members covering more than 55 countries and 50
industries, EFQM aspires to provide a ‘platform’ for organisations to learn from each
other and improve performance. EFQM is the custodian of the EFQM Excellence
Model, a business model whose aim is to generate “Sustainable Excellence”. The
model has three components: (eight) fundamental concepts of excellence1, the ‘9
The ‘fundamental concepts are: “taking responsibility for a sustainable future,
Achieving balanced results, adding value for customers, leading with vision,
inspiration and integrity, managing by processes, succeeding through people,
nurturing creativity and innovations, building partnerships” (EFQM web site,
accessed November 2011)
1
10
Criteria’, and the RADAR (results, approaches, deployments, assessments and
refinements) logic.
The core is the cause and effect model between five enablers and four types of results.
Insert Fig 1
As with many such models (such as the BSC) the precise make up of the model is
neither standard nor prescriptive in its specifics, and like the BSC has a common aim
is helping organizations define and then implement a strategy in terms of the elated
processes and ‘balanced results’ designed to achieve the strategy.
Balanced Scorecard (BSC): The ubiquitous BSC is also a model viewed as a major
component of the divisions processes that linked to strategy. As commonly known the
BSC has the four quadrants: customer/market; financial; internal processes; and
people/innovation.
Insert Fig 2
Each quadrant is analysed in term of goals, targets, action plans and measures thatare
intended to help implement firm strategy and feed that strategy downwards
throughout the organization in a cause and effect relation. Like the EFQM Excellence
Model, the precise targets, measures, and the specifics of ‘balance’ are a matter for
the individual organization to discern, and in that respect it is also non-prescriptive.
The BSC has historically had a training arm, the Balanced Scorecard Collaborative,
now Palladium.
Process Improvement (PI): In contrast to the EFQM, PI is an internally derived
system within the Division to which Automata belongs. The annual policy and
strategy review undertaken in Automata combined with the BSC generation are
assumed to be “the principle mechanisms for identifying key business processes and
the associated metrics to demonstrate control and progress against them” (Internal
Document, 2004). Another Internal Document (2003, emphases in original) states
“processes are identified in and form part of the Mission statement for the business”.
This is summarized in the catch phrases ‘Proven choice for manufacturing’ and
‘Aiming for excellence’. The PI is a benchmarked best practice approach.
Graphically, it is represented as a set of seven blocks representing a Division’s
Business Process. The seven blocks are organized in the shape of two side blocks
11
designated on the left ‘Process Management Methods’ and on the right Process
Management Roles and Responsibilities. In between these two blocks, five blocks are
stacked on top of each other: Management Process; Product Life Cycle; Supply
Chain; Customer Relationship; and Support Systems. Evidently, Management
Processes and Supply Processes are seen as processes that would facilitate the
attainment of the three processes in-between. The process framework could also be
disaggregated into any specific unit process – e.g. ‘Customer Focus’, ‘Gigantic
Competitiveness’ or ‘Innovation’.
Insert Fig 3 here
For each of the five processes stacked in the middle, one or more managers are
designated as owner(s), and each process is split into a number of specific functions
that are aligned with specific measures. For example, the MD, the Commercial
Manager, and the Quality Manager jointly own (are responsible for) the Management
Process. Examples of its responsibilities are communication and budgeting, for which
information is measured in the EFQM, BSC, team brief data, and sample feedback.
Supply Chain is owned singly by the Manufacturing Manager, and its responsibilities,
for example, include reduction in assets tied up, delivery performance, material on
time delivery and process cycle time, and its measures are drawn from the BSC. As a
final example, the HRM Manager, the Purchasing Manager, the Manufacturing
Engineering Manager, and the Quality Manager jointly own the Support Process. Its
responsibilities include recruitment and development, health and safety, and internal
audits, and these are measured through training days per employee, staff turnover,
sickness, number of accidents, and number of risk assessments.
Time: Time is another internal model aimed at strategic cost control, increasing
profitability by stimulating sales revenue and reducing costs across the whole
organization. To achieve these ends, Time relies heavily on innovation and
benchmarking, so the focus is upon efficiency and productivity. Each business (such
as Automata) within Gigantic has discretion to use Time in whatever way it deems fit
to its business and market activities:
“It’s not prescriptive inasmuch as it doesn’t say that you have to do this and
you have to do that. It says you have to look into these areas, but it’s up to you
how you go about it” (Manager).
12
Time, in certain regards perhaps reads most like a disaggregation of the BSC
quadrants. ‘Success’ is premised on the capacity to disaggregate inputs and outputs
into goals measures and consequences. The system overlaps with Process
Improvement also in that each model of goal, measures, etc is related to innovation,
customer focus and global competitiveness in the seven-block model. By relating
costs to innovations, competitiveness and customers the model attempts to join up
initiatives in terms of their multiple consequences within the organization or
organizational sub-unit. The Time basics of innovation, customer centrism and global
competitiveness attempt to shape processes such that external benchmarks are used as
reference points for the goals, measures and results that influence internal cost
reductions and profitability improvements – however achieved.
Insert Fig 4 here
Like Process Improvement, Time was an in-house company model, established within
Gigantic that offered another linear process model of the relationships between
strategic company programmes/goals and divisional strategy implementation.
Having outlined the four management control systems in the next section we move on
to consider the workings, practical rationales and interactions that each system
seemed to display within Automata.
The Work of Multiple Management Systems
We start with EFQM because this was the measurement system our informants
discussed most frequently, and which they considered the most relevant to their
everyday work as divisional unit managers. The EFQM was the earlier of the two
external performance measurement systems (PMMs) used by Automata. It was
introduced within the divisional unit in the 1990s as the rating method of the total
business and to analyze key processes during what was a turbulent time. EFQM was
also the choice of the Automata staff and strongly associated with their attempts to
operate as ‘business unit’. To understand this we need briefly to describe Automata’s
recent history.
Automata’s experience had been volatile, having had four Managing Directors during
1996-2004, “each one having different leadership styles” (EFQM Report, 2004). It
began as a manufacturing workshop facility and a sales office. In the 1980s, it was a
13
profitable business, but as its major customer began to shop around for the cheapest
supplier, Automata was left with sizeable spare capacity. To overcome this problem,
Automata diversified into subcontracting, which eventually outgrew the core product.
This, however, did not meet with Gigantic’s approval, since it wanted Automata to
concentrate on manufacturing rather than subcontracting. This intervention by
Gigantic led to the reconfiguration of Automata as a ‘factory’, attached to a large
manufacturing division with the mission to develop a “world centre of competence”
in standard electronic products. Automata was to focus on high volume, low
complexity products to exploit its available capacity and the expertise gained while
working as a subcontractor. This decision was taken by Automata staff as a major
positive signal
“To get that factory volume was quite a big thing for us and really the story
has been a roller coaster from then on.” (Manager).
Hence, by the early 1990s, Automata had changed from “a basic metal bashing
activity into a very high tech, clean environmental assembly area” (Manager), albeit
within a very cost conscious strategy. By the mid-2000s Automata was increasingly
viewed as a success story with Gigantic.
From the mid-1990s, Automata was designated a semi-independent business and
given increased scope for autonomous action for two years during which it functioned
“like clockwork” (Manager), producing £5M profit to HQ each year as it had control
over manufacturing, sales, design, and marketing. However, a reorganization was
prompted by HQ because of difficulties faced by one of its major divisions, resulting
in centralizing sales, which took Automata “back again into the days of the factory
with the design activity attached” (Manager).
Within the re-organization, Gigantic determined the transfer price for Automata’s
products using a top-down calculation. The process began with customer price in the
field, which was then taken down to sales prices (as offered by Gigantic sales units)
and from which gross margin was deducted to arrive at an ex-works transfer price
(meant to cover full works costs). A differential transfer pricing system was used
whereby high power rate products were making a profit margin of up to 40% whereas
low power rate products had a transfer price of cost minus 10%, and hence the
success of Automata depended largely on product mix. Moreover, a significant
14
decline (by more than one third) in the value of one European currency against
Sterling worked against the financial performance of Automata: “despite our volume
in pieces growing our volume in turnover wasn’t growing. So consequently our
profitability was getting hit very hard because of the exchange rate” (Manager). This
prompted HQ to issue an ultimatum that unless Automata turned these figures around,
it would be shut.
Managers recalled that period as one in which employees lacked any sense of
ownership in the plant, and management had no strategic vision for the unit nor
careful production plans. While Automata’s products had willing customers, the plant
took the view that it should overproduce in order to improve productivity. This
proved costly to the plant: when there was a downturn in sales, Automata was left
with “a massive risk” because the design was not stable, leading to frequent recall of
faulty products from customers and making the necessary corrections in all units of
inventory left over.
Automata’s senior management recognized the seriousness of this pressure:
“The pressure [from HQ] has always come back on the factory, ‘you’ve got to
do more, you’ve got to do more, you’ve got to do more’ and we
responded…because if we hadn’t responded we would have been shut down...
without any question whatsoever” (Manager).
Automata’s managers now had to justify its survival by being benchmarked against
the cheapest producer internationally:
“Along came China and we were literally told two years ago ‘guys you either
meet the cost level that China can offer or you know it could be curtains”
(Manager).
Further, HQ began negotiating with a Japanese company the possibility of
outsourcing the AMANDA (pseudonym) project whose aim was generate future
products that would replace existing products.
To deal with these challenges, the strategy of Automata managers was to convince
HQ that they knew what “costs are for all our product lines” and hence they should
“make money or at least break even on every product line” (Manager). This, however,
required that HQ refrains from manipulating transfer prices by following a straight
pricing strategy across the whole product lines, and thus for HQ, rather than
15
Automata, to take the risk of making profits or losses on product in the market.
Automata managers argued successfully that Gigantic should use a bottoms-up
approach so that the transfer price is insensitive to variations in power rating but
simply adds a fixed margin percentage to ex-works costs. Gigantic countered by
producing an alternative transfer pricing system to produce an equalization effect, so
that Automata would exactly breakeven, so that:
“Whatever gains that we were making in the manufacturing side tended to be
passed on to the sales with a view of growing market share. So the pressure on
manufacturing is always there, you earn it one year, you give it away at the
start of the next year and you have to start again” (Manager).
Automata’s managers:
“turned around and said ‘we’re not here to play around with transfer prices
because if I increase the transfer price by three hundred per cent it doesn’t tell
me that somebody in the factory is working more efficiently or we’re getting
better productivity” (Manager).
The aim of Automata therefore was to secure a stable transfer price that does not get
adjusted every year to swallow the efficiency savings it made. This they have been
successful in securing, which led the management of Automata to concentrate on
efficiency savings. This change, coupled with a more favourable exchange rate
movements more recently means that Automata has begun to generate “something in
the region of about two million pounds net productivity a year, cost productivity
reduction per years in those elements of design to cost, material cost and process”
(Manager). These improvements paid off, as HQ called off the deal that was being
negotiated with the Japanese company, and Automata remained in operation. In short,
Automata has benefitted from a greater focus on processes and associated costs.
As part of the changes to encourage ‘improvements’ in processes and design, one MD
introduced EFQM during 1995. This MD was in favour of EFQM as he believed the
EFQM reflected a more engineering approach than rivals such as the BSC – which
were viewed as more ‘accounting’ or financial in orientation (Interview, Manager).
This allied with the new MDs own engineering background. Automata then developed
their own system of balances using the underlying EFQM Business Excellence Model.
It is divided into two sets, each weighted 50%: enablers and results, and these scores
are then allocated in precise ratios among the themes that make each set. The enablers
include inspired leadership (10%); incorporation of quality values and concepts in
16
policy and strategy (8%); releasing employees’ full potential through people
management (9%); providing necessary resources, including financial, material and
technologies (9%); and reviewing and revising processes (14%). The results set
includes enhanced satisfaction of employees (9%); customer satisfaction (20%); better
impact on society (6%); and improved business results (15%) (see Ezzamel et. al.,
2004).
By 1997 Automata was fully commited into using EFQM and had taken up
membership of the EFQM association, which offered feedback, assessment and
inspection services, on a regional basis. And as we noted above EFQM was very
strongly linked to the improvement programme.
The Accretion and Overlapping of further Measurement Systems
However, in 1999 Gigantic HQ introduced the BSC into Automata and Mercado as a
result of a parent initiative to roll it out throughout the whole group, although our
informants have suggested that the BSC had been introduced elsewhere in the parent
company years earlier. It has been developed in Automata, it cascades down to the
level of each department. In Automata’s EFQM annual document, the BSC is
depicted as a major component that is directly linked to the plant’s strategy and feeds
into the annual review, which in turn feeds into strategy, and so the cycle continues.
The BSC has the familiar four quadrants: customer/market; financial; internal
processes; and people/innovation. One Internal Document (2003) states that “All
measurable business performance data is recorded on the Balanced Score card”. The
EFQM documents (2004) states that “Some 40 individual measures exist on the Top
level Business Balanced Score card. These measures are considered the key indicators
for the ‘health of the business’. Each measure has a target that reflects either agreed
business goals or stretch goals to achieve improvement…. In addition, a ‘target for
excellence’ that is considered to be a truly world class target which should be
expected by companies operating within our field is also shown on the Balanced
Score card for each measure.”
It is claimed that actual performance is monitored against the relevant measure/target
depicted on the BSC monthly at the Process Managers’ Meeting and is also used as a
17
key input to the Policy & Strategy Review. At that Review, the measures stated on the
BSC are checked for relevance and changed/replaced accordingly, and new targets are
agreed and set. Yet the reception of the Balanced Scorecard at Automata was at best
‘lukewarm’. The KPIs identified as key to the four perspectives are calculated but
these same KPIs are routinely conceptualized as the same KPIs as calculated for the
EFQM Business Excellence Model, which was consistently favoured by managers as
more ‘improving’ than BSC. These common KPIs were also the basis for the
division’s incentive compensation scheme for management.
As one manager noted:
“Our BSC doesn’t really mean an awful lot to most people… at the
[Divisional] level we’ve got a BSC but we don’t ever review it. We have the
charts in our management information pack which people look at and you
know take some comfort from or otherwise, but we’ve never actually
reviewed its content and said are we still measuring the right things. It just
kind of exists… What is the point of having a metric that no-one ever uses?”
(Manager, Mercado)
The BSC was viewed as something that “doesn’t really mean an awful lot to most
people” could be that it is not considered as adding to the information or measurement
process. As one Mercado manager stated: “the BSC exists, it has data, but as a
meaningful indicator of how well [the Division] is performing I don’t think it’s all
that meaningful.” The questioning of the importance or extra value of the BSC was
reflected in the way that managers could speak of the interrelations or overlap
between the EFQM and the BSC.
Part of the reason for this downgrading of BSC seemed to be that managers
considered EFQM both sufficient and effectively covering the measurement domains
of BSC (and more). Our informants in Automata frequently referred to the
multiplicity of management systems circulating at work. Many saw this multiplicity
in terms of overlap, or overlay of parts of one system upon another. In commenting on
the multiplicity of measures at Automata, one manager said
“I’m not sure that we need all, for example the BSC and EFQM, there’s
overlap between those two.”
For this manager, the BSC and EFQM almost exactly overlay on each other to the
extent of redundancy, hence suggesting that there is no need for both. Similarly:
18
“The results side of the EFQM model does really measure the effectiveness of
what’s on the left hand side, and the BSC is measuring the effectiveness at the
higher level of the organization. The two are the same.” (Manager, Automata)
The issue of system overlap, however, becomes even more salient when Automata
was required to introduce the Time system that had been trialled in Gigantic’s other
units during the late 1990s.
In 2000, after Automata was granted semi-autonomous business status and granted
further strategic autonomy within Gigantic, Gigantic’s HQ started to roll-out its own
process improvement system that all divisions were required to implement. Time and
Process Improvement were joint initiatives designed to address business improvement
programmes and plot design initiatives and changes in ways strikingly similar to
EFQM. The rhetoric of the two (Customer Focus, Innovation, Competitiveness) and
the linking to specific design channels or programmes with specific goals and
measures matched significantly the linear process, categories and criteria of the
EFQM Business Excellence Model. As one manager reported, any distinctions
between the two, which often counted for little in everyday work, had commonly to
be explained to staff within Automata and Mercado:
“I had somebody in here yesterday saying, ‘explain this [Time] and EFQM
thing because we’ve got two systems’…. People get a bit hung up on it by
saying ‘should we do an EFQM or should we be doing [Time]’? And really if
you’re doing EFQM, you’re doing [Time]… So it’s an on-going story.”
(Manager, Mercado)
In these quotes, working with the BSC and EFQM, or Time and EFQM is assumed to
lead to the same results, so that the two systems are assumed to be almost identical in
their effects. Combining the three quotes above would seem to suggest that there is
little difference between the BSC, EFQM and Time. These three systems have been
presented in these quotes as if they were overlapping or nested systems, without
suggestion collision, conflict or clashes. Some differences were recognized but for the
managers working with EFQM they were considered minor:
“[Time] …has some structure to it, some toolsets, some things that you can
use to help you identify opportunities for improvement. And that’s where it
starts overlapping a little bit with EFQM ‘cos EFQM has got some of that sort
of stuff as well.” (Manager, Mercado)
19
In which case, given the similarities or complementarities, and lack of overt conflict
or contradiction between the two, why did Automata persist with EFQM? Or
alternatively, why did Gigantic require their divisions all to use the Balanced
Scorecard, Time and Process Improvement?
Multiple Management Systems and their Recognition
The interviews suggested to us that Automata had a substantial pride in its
implementation of EFQM through a difficult period for the business and through
which it negotiated successfully. So spectacular was the turnaround of Automata
considered to be that within Gigantic, “they now whenever they quote anything as far
as manufacturing is concerned they refer to [Automata] as being the Benchmark”
(Manager), as the site was shown to be the most cost effective manufacturer (EFQM
Report, 2003). Yet, amid this euphoria, the message from HQ for the immediate
future was not necessarily secure. During the congratulating visit by a senior HQ
manager to Automata, after praising staff for their remarkable achievement, he said:
“‘Productivity is your life insurance.’ That had more of an impact on the
people sitting round the room than the previous two hours of congratulations,
because he told us that you may be successful today but once I go away and if
you stop being successful the situation will change. So we know we can’t sit
back and relax… because everybody else is going to be working twice as hard
to try to knock us off that position” (Manager).
This was also expressed in formal documents: “The ability of [Automata] to deliver
annual productivity gains will decide its future… Continuous improvement driven by
the excellence model assessment is our way of managing our future” (EFQM Report,
2003). These aspirations are reflected in Automata’s declared mission and strategy,
which emphasize growth, productivity through continuous improvements, quality, and
customer focus.
Moreover, an important element of the Automata’s turnaround was considered by
their own managers to be down to the implementation of the EFQM Business
Excellence Model. EFQM as we noted previously operates as a membership
association that offers advice between companies, benchmark data and consulting
assessments. It has some attributes similar to the BSC Collaborative with
memberships, advice, and awards for successful implementation. Yet Automata’s
membership of the EFQM was on a regional rather than European basis; Gigantic was
20
not a member. Within EFQM, there are various levels of ‘recognition’ for the
programmes member firms have introduced. The ‘Excellence Award’ is based on
documentary submission to the EFQM, followed by a five hundred hour assessment
visit by an expert assessor team. Each year EFQM nominated a member for a
competitive award for Excellence based upon the assessors’ visits. For the Balanced
Scorecard, the equivalent would be the Collaborative’s ‘Hall of Fame’ (Ittner, 2007).
From 1998, Automata had prepared self-assessment documents to submit for
Excellence Award and had received it on multiple occasions. As well as this
recognition, considered prestigious by Automata’s management, the assessor team
from the EFQM also provided an improvement report with relevant data for
benchmarking from other EFQM members (over 500 corporations).
“[That is] why EFQMs quite good actually, because we get information that’s
outside our industry” (Manager, Automata).
The premier motto of the EFQM is “Share what works” (EFQM:
http://www.efqm.org/en/Default.aspx). In this way not only did Automata’s
management take pride in their turnaround of the divisional unit, but also attributed it
in part to their success, externally validated, with a business improvement/excellence
model and the feedback they received from its implementation.
However, Gigantic’s HQ were known for their antipathy to EFQM, and this attitude
seemed to derive from several issues. First, Gigantic had disclosed a lack of trust in
EFQM’s measurements borne of internal events. A few years prior to Automata’s
membership of EFQM, we were told, two other European businesses within Gigantic
won EFQM awards, but then one year later ran into serious financial difficulties. This
demanded some explanation:
“someone at a higher level had some egg on their face having promoted the
fact that ‘look how good we are’ [based on the EFQM awards], and then next
thing you know is the news from a financial point of view and had some
explaining to do… it’s a case of once bitten, twice shy [for them]” (Manager,
Automata).
Automata’s continued success, recognised within Gigantic, in short, has never been
acknowledged by the HQ to be related to its use of the EFQM Business Excellence
Model. This apparent lack of credit, which is comprehensible also in the sense that it
21
is difficult to attach definitively performance improvements unambiguously to any
performance model, did not play well with Automata’s managers.
Second, there were other explanations, relating to Gigantic playing out identity issues
with other major corporations and the sponsors of EFQM. Managers within Gigantic
and Automata both reported that HQ staff had also a longer standing antipathy
towards EFQM that derived from the genesis of the EFQM foundation. EFQM was
founded in 1988 as a non-profit association of premier European Corporations that
came together to develop and promote an Excellence Model for European businesses.
Gigantic was never part of this founding group, and many managers within Gigantic
had remarked on this. Whether this was the choice of Gigantic, or a question of the
founding members ignoring Gigantic to maintain their own status and prestige within
the emerging Foundation is not clear. The Gigantic Corporation at HQ level has
continued to maintain a distance between EFQM’s self-styled ‘European Model of
Excellence’ and its own corporate wide management systems.
This antipathy between EFQM and Gigantic, appeared to shape much of the approach
that Automata’s managers took towards Gigantic’s own management systems – in
particular towards Time. Noting the similarities between Time and EFQM was at
once an acknowledgement that there was significant similarity in key concepts and
general structure, but also a way of criticizing implicitly the requirement to ‘foist’
Time onto Automata when what they had already seemed to them to be working well.
Several informants spoke of the parent’s drive to assert its identity through the
development of its own initiatives that imitate the performance management systems
available in the business world, but repackaging them as a special brand of the parent:
“When Time was first introduced it was introduced as [Gigantic’s] business
excellence program. And the significant word there was [Gigantic] as all we’d
done really was take EFQM out and put [Time] in, so because [Global] is big
enough and ugly enough to do its own thing it did, instead of saying, OK
we’ve evaluated all the different tools around for improving the business and
we think EFQM is the way to go, therefore, [Gigantic] is going to buy
wholeheartedly into EFQM…. I suppose it was just typical of [Gigantic’s]
arrogance to things we’re using.” (Manager, Mercado; emphasis added)
Other managers were also critical of Time in comparison to their familiar
management systems:
22
“We’ve got a [Time] toolset, and these were meant to be tools that you used to
try and tease out the improvement opportunities. But it just didn’t translate
very well to English, and so I looked at them and just remember coming away
totally confused by what the hell they were supposed to be used for… It lost
something in the translation, and I remember looking at it in the early days of
me doing this job thinking, ‘well is this the tool set I need to drive the change’.
And I just remember looking at it thinking, ‘I don’t understand this, I don’t
really know what I’m supposed to be doing with it’. And from my involvement
with other people in jobs like mine they all basically feel the same way.”
Manager, Automata; emphasis added)
If EFQM seemed to be tied to Automata’s identity and recognition of its
achievements, then Time seems similarly to be an initiative that was tied to the
identity of Gigantic HQ; rather than simply buying into an existing management
systems, such as EFQM, Gigantic’s identity is thought to be affirmed and promoted
by a repackaging of EFQM into Time. Of course this is only one possible reading of
Time; another reading would one that emphasizes an organic development of Time as
a system tied specifically to the context and needs of Gigantic, as suggested to us by a
senior manager at the parent’s HQ: Gigantic can now parade Time as its own brand of
management system. Yet the above reasoning is compatible with playing out of
identity issues by Gigantic, as is the possibility that specific management systems
could be promoted as a statement of independence, in the face of perceptions by the
company of being ignored by the EFQM business grouping:
“Someone once told me ‘well the reason why [Gigantic] never pursued EFQM
as a global initiative is because the fourteen companies or heads that formed
the original EFQM group wasn’t represented by [Gigantic].’ If you were to list
the top fourteen European companies I think [Gigantic] would be in there. So
for some reason, why weren’t they in there and may be someone was a bit
offended that [Gigantic] weren’t involved and thought, well OK we’ll do our
own thing.” (Manager, Mercado)
One quality manager, from Automata, also remarked:
“[TIME] is a [Gigantic] improvement initiative, and I’ll make a comparison
with GE if I could. GE are pushing Six Sigma as their total business
improvement [model], [Gigantic] I would suggest are pushing [Time]. They’re
similar methodologies, they’re all about not being satisfied with the norm and
having structured approaches to improved business performance… I think
they [HQ] wanted [Time] to be part of their identity, they wanted financial
institutions to realize they have an improvement program.”
These multiple explanations offer a glimpse of the complexity faced by informants
when attempting to justify or rationalize why certain management systems are
23
favoured, or marginalized, or even ignored altogether. But they also speak to issues of
status and recognition that managers were sensitive to in explaining their choices and
those of others with respect to their favoured management system.
In the final section of the case we discuss how the operation of multiple management
systems functioned between different units within Gigantic.
Working Across Multiple Management Systems: Accommodation and Conformity
In the previous section we saw how relationships between the centre and its
environment, and the divisional unit and the centre are major undercurrents in
comprehending how multiple measurement systems can co-exist, if not necessarily
co-function. The organizational etiquette of working across or handling such tensions
is also intriguing, and we noted that it took on a variety of forms. One form would be
to avoid making any reference to performance management systems developed
locally which may be considered offensive to the parent:
“I do believe that when some of our people go to [parent HQ] and they get
involved in these types of discussion they keep it pretty low key, the EFQM
activity in the UK, simply because in [HQ] it’s not necessarily smiled upon as
an improvement tool. They would rather talk about [Time] than EFQM.”
(Manager, Automata)
Managers consulting or meeting with HQ managers would tend to make no explicit
mention of their use of EFQM, knowing that only the latter was approved officially.
An alternative style might be to ‘talk the talk’; that is to speak the management
control language the HQ prefers to hear:
“I think when we’ve had very senior heads [here] we’ve made sure that we’ve
got our [Time] badges on, and as soft as that [is], I mean we’re playing the
game.”
What is emerging from these empirical details is a situation where parent initiatives
are viewed as simply a badge, or even totally irrelevant, at local levels.
But, what about the units geographically closer to the international HQ? Our
interviews suggest that such initiatives are much more explicitly promoted at the
international HQ level and in units close to it. Geographical distance between
24
Automata and Gigantic HQ seemed to lessen the oversight Gigantic gave. In response
to the question “In day-to-day activities does [Time] figure out in your thinking
explicitly?” one manager (Automata) responded:
“No, I think it’s a logo. If you were in the (X) factory [near international HQ],
and you were asking that question you’d get a very different answer, you’d get
it that [Time] probably guides all of their improvement approaches.”
This suggested that in terms of centralizing and decentralizing rationales of Gigantic’s
divisionalization logic, the greater the de-centralization the further the divisional unit
was from the geographical place of the corporate HQs.
More importantly, however, we noted how the managers at Automata worked across
the management control divide, such as as it was, between HQ’s preference for Time
and BSC, and their own preference for EFQM using a strategy of accommodation.
‘Accommodation’ here refers to the linguistic act of modifying speech and talk to the
language or dialect of another person outside of your immediate context (Miller,
2005). Talk Time to Gigantic HQ, talk EFQM to your immediate colleague. As we
noted earlier the homologies between the categories of the two systems made this
kind of cross-over relatively unproblematic, and our informants noted that it was a
skill at which they were well practiced. As one manager noted:
“[Time] is quite [country-specific] in its structure and in its objectives.”
Managers at Automata, however, were keen to express concern that the
rapprochement between HQ systems and their unit’s systems might be ending. And,
as we would expect from rational economic theory, the question of cost of multiple
systems was at the centre of these concerns.
Gigantic HQ has recently been viewed by local managers to be ‘flexing its muscle’,
with a new change towards achieving greater homogeneity of practice across the
organization worldwide. This is manifest, for example, in a recent (2007) drive to
ensure that all local units embrace Time:
“[Things] are changing now and [Time] is being mandated by [the] UK CEO,
that’s all business improvement activities in [Gigantic] in the UK across all
activities have to be branded as [Time]” (Manager, Mercado).
Local units now are asked to submit periodic reports on their objectives on the Time
initiative, what it means, and what improvements have been attained (Manager,
25
Electron). To reinforce this focus, the Gigantic’s in-house magazine published regular
articles on Time so that: “you can see it starting to permeate” (Manager, Automata).
The assessment documents produced by Gigantic demand transparency, clear
responsibility, measurement, and process understanding. This seems to be pushing the
local businesses within the group towards more compliance:
“we felt that we needed to push processes a little bit more for a number of
reasons, one we can improve our scores on excellence assessments, two we
get more people talking about processes, three we tend to push towards
compliance with unwritten [Gigantic] edicts, you know their good practices,
and four the requirement of ISO9000 2000 was one where you needed
transparency of processes” (Manager, Automata).
Also, a new web publishing technology has been actively promoted by Gigantic,
which,
“can be from one extreme a very simple flow chart to a full blown
demonstration of compliance linking all the documents that we use within the
business” (Manager, Automata).
For example, one-day seminars on Six Sigma have been launched by Gigantic where
managers from local units were expected to attend, with the aim of enhancing
compliance within the group worldwide. One Manager (Automata) noted,
“It’s one day seminar yes and again you can see that’s also a little bit of
political pressure …. It would have been subtly pushing us towards that
conformity, you know the e-mail that invites you doesn’t say you shall go.”
Failure to attend these seminars by local managers, and inability to articulate a clear
response when attending to show that the local site treats Time as part of the site
strategy would be taken by HQ managers as negative signs.
“more and more, we’re getting the influence of [Time] and things like that are
coming through, yeah. Another example would be a thing called [Project
Spirit] within this company now. What’s happening is that across the whole
world [Gigantic] is trying to harmonize its processes, and the way it’s
structured, so in Northern European countries [sites] are doing this. So right
now we’re engaged with the likes of Holland and Finland and Sweden and
Ireland to agree common processes. Now you think about that, that’s just so
difficult, I mean you can’t even imagine how you’d go about doing that. But
nevertheless we’re doing it, and the reasons are that if we get common
platforms we reduce them for variations, we can save a lot of money.”
(Manager, Mercado)
26
This pressure for the group to become globally competitive may be driven partly by
the desire to cut costs down as indicated above, although whether the cut costs idea is
a product of rational calculation of the costs and benefits of each system is unclear.
This partitioning of the world by Gigantic in the name of conformity is beginning to
be seen by divisional as making a difference in the sense that greater centralization is
felt to be closing in on them. Managing in the name of greater conformity is seen by
the local businesses to be threatening to their identity, disruptive to best practices that
they have developed locally, and costly to them, even though it may produce
significant savings for Gigantic:
“It’s coming straight down the track, it’s a bit of a scary thought really
because what’s going to happen is that right now, because we’ve had SAP
since about 1992, so we’re talking twelve years we’ve had two or three
versions of SAP in that time and we’ve become good with it, it meets our
business needs… And it does what we want, and now just to make an extreme
example, we have a very good SAP returns process, which we think is best in
class. Now [if] in harmonizing these processes we have to employ four more
people to do the extra work this doesn’t make us more competitive. But the
answer is that right at the higher level in the organization this is expected to
save us tens of millions of pounds, so it’s worth doing at that level, we just
have to live with the consequences.” (Manager, Automata)
Another concern raised by local businesses in the face of the new sustained drive by
the parent to secure compliance is the loss of local autonomy, and with it inevitably
own independence:
“Once we [harmonize] what we’ve got is what we’ve got, and if we want to
make a change then ten other countries have got to agree, it’s a bit like the EU.
So it’s a worrying thought really ‘cos you know we’ll have lost a lot of the
flexibility.” (Manager, Mercado)
HQ-based initiatives, such as Time, and to a lesser extent the BSC, seem to be acting
as forces seeking to draw remote sites in the direction of international HQ, and these
seem to be having some impact, particularly when the local sites are within
geographical proximity to HQ. Locally-selected systems, such as EFQM, increase the
distance of local sites from international HQ, and by implication enhancing local
autonomy and identity. This centre-local interaction is a dynamic process with the
force of centralization and decentralization waxing and waning across time and space.
Yet, although our interviews with Gigantic, Mercado and Automata concluded in
2007, in 2012 we noted that Automata continued to have membership of the EFQM,
27
and continued to be recognized as successful in its implementation of the Business
Excellence Model. Fears of centralization that managers at Automata reported to us,
seemed to have remained fears, and the rationale of cutting costs by reducing the
multiplicity of management systems seems thus far not to have been followed
through.
Concluding Discussion
Three concepts seem central to the functioning of multiple management systems in
organizations. The first, which is the subject of much of the research on the role of
performance measurement and management systems, refers to their ‘work’ in the
sense that the systems work and are worked as tools of monitoring, analysis and
decision-making. In the case of Gigantic, EFQM, BSC, PI and Time were all
deployed as monitoring, measuring and evaluation tools, but not equally inside the
units of the organization that we examined. Each system had an active role for
monitoring and evaluation within the organization for some. For sub-units or the
centre of Gigantic several but not all the management systems were deemed useful
and informed interactions between managers, units and the central HQ.
The work of these management systems however, depended upon a degree of
interchangeability between their terms, frames and categories. Of the four systems we
explored the BSC was considered the most compliant – mainly as a result of its
seemingly inherent malleability, a point critics have noted previously (Busco &
Quattrone, 2010). Automata staff found little difficulty in nesting BSC within a
Business Excellence framework. This gave BSC possibly the least contentious profile
of any of the systems. However, we clearly noted differences in usage regarding the
process improvement management systems. Automata strongly favoured EFQM, a
system to which the central HQ had limited tolerance. Gigantic’s desire to develop
their own system encountered antipathy from Automata staff that had been using
EFQM for almost five years already, with apparent success. We suggest that the
continuing commitment of Automata staff to EFQM is path dependent in that it is
founded and reinforced by their historical trajectory from a struggling ‘factory’ to
successful semi-autonomous business within Gigantic. Willingness to shift
management systems was undermined by the apparent historical experience and
28
accomplishment of using EFQM. What sustained the multiple systems were their
overlapping terms, whatever their different governing diagrams and figures.
This leads to our first research proposition:
R1: Where measures and goals overlap between systems, the use and
effectiveness of ‘branded’ or proprietary management systems cannot be
determined independently of pre-existing systems.
Yet the paths that Gigantic took in arriving at a point where sub-units use multiple
systems also reveals something of the desire for identity and recognition that
organizations and their autonomous sub-units seek from organizational structures and
management systems. Whether the EFQM was functionally effective, or in some
ways superior, to other improvement systems (such as Time or Process
Improvement), it is clear that Automata’s reluctance to move away to Gigantic’s
internal system is related to their positive experience with EFQM (successful), and the
recognition that EFQM Awards had brought to their endeavours. It should also be
noted that this recognition was also not simply legitimation, but the feedback received
from the award assessment process was regarded as a useful source of both feedback
and benchmark data from other members – suggesting ‘rational information’
justifications also. Further, issues of recognition are at the same matters of identity
and of being recognized as successful implementers of the Business Excellence
model. Such recognition helped to reinforce Automata’s commitment to the system.
Recognition can also been seen in the stance of Gigantic’s central HQ’s reluctance to
embrace EFQM. Gigantic HQs lack of involvement with EFQM at the founding
stage, and the apparent ambiguous response they now had to the Award process,
given that other sub-units had been awarded and then struggled financially, shaped
corporate disapproval to the management system. Yet the desire for recognition
shaped Gigantic’s decision to develop their own management process systems and to
implement them throughout the corporation. The momentum to develop Time and
Process Improvement illustrates some of the norms that senior management perceive
as appropriate features, systems and structures of a prestigious corporation. Similarly,
Gigantic’s reluctance to embrace EFQM was also suggested to be connected to their
unwillingness to join the foundation once the model had been formulated – rather than
during its development by the fourteen founding members.
29
This leads to our second research proposition:
R2: The use and effectiveness of any specific management control system
is intertwined with the recognition that a management system grants to its
users and the identities that managers and other users bring to and take
from their use.
The continuing existence of the multiple management systems also owes much to the
capacity of Automata staff to speak the language of Time and PI when dealing with
corporate headquarters. We describe this as consonant with the process of
‘accommodation’ found in linguistics when speakers across languages or dialects
adapt their own speaking voice, vocabulary, pronunciation, etc., to each others way of
talking in their interactions. Most notably Automata openly described ‘talking the
talk’ of time and BSC when interacting with Gigantic HQ staff. The approved
language of Time and BSC replaced all reference to usage of the local dialect of
EFQM.
This leads to a third research proposition:
R3: Management systems only co-exist to the extent that users within
organizations are able to accommodate each other’s performance
measurement languages and concepts.
Finally, we noted the stirrings of a questioning that we expected to find much more
prevalent within Gigantic over the four years of our study: isn’t the existence of
multiple management systems cost ineffective, or extravagant even? Gigantic HQ was
starting to inquire whether it might be more efficient to standardize and harmonize
management processes. We noted also, however, awkward questionings. Is not the
imposition of the controlling management system of performance measurement
moving away from the kind of autonomy that divisionalization is supposed to foster –
local preferences, initiative, semi-autonomous profit centres etc. Moreover, four years
after we last conducted our research interviews at Gigantic, Automata is still using
EFQM within its division.
From the concepts of work, recognition and accommodation that supported the
existence and persistence of four management and measurement systems within
Gigantic, we finally draw out research proposals and propositions for further study.
30
Our initial literature review noted some of the divergence of research styles into the
uses and consequences of management systems within organizations. One clear
implication of our study is that within organizations there might exist some variety of
systems located at different levels and geographical sites within large corporations. It
is mistaken to assume that ‘organizations’ have a coherent management system or
even more than one system that they necessarily deploy in a uniform style (Dearden,
1972). This suggests we need to more carefully consider what it also means to talk
about a ‘control package’, since this term also implies some notion of coherence or
complementarity.
Instead, we suggest that commitment to a management system will be structured by
the historical conditions of its diffusion within the firm and the formative experiences
with its use. The historical channels of diffusion are the clue to issues of use,
recognition and identity that shape managers preferences for particular management
systems (Strang & Meyer, 2003).
31
References
Abrahamson, E. 1996. Management fashion. Academy of Management Review, Vol. 21 (1), pp. 254285.
Abrahamson, E. 1991. Managerial fads and fashions: the diffusion and rejection of innovations.
Academy of Management Review, Vol. 16 (3), pp. 586-612.
Abrahamson, E., & Eisenman, M. (2008). Employee-management Techniques: Transient Fads or
Trending Fashions? Administrative Science Quarterly, 53(4), 719-744.
Baiman, S., Netessine, S., & Saouma, R. (2010). Informativeness, Incentive Compensation, and the
Choice of Inventory Buffer. The Accounting Review, 85(6), 1839.
Banker, R. D., Chang, H., & Pizzini, M. J. (2004). The balanced scorecard: judgmental effects of
performance measures linked to strategy. Accounting Review, 79(1), 1–23.
Campbell, D., Datar, S. M., & Sandino, T. (2009). Organizational Design and Control across Multiple
Markets : The Case of Franchising in the Convenience Store Industry. The Accounting Review,
84(6), 1749-1779.
Chenhall, R.H., 2003. Management control systems design within its organizational context: findings
from contingency-based research and directions for the future. Accounting Organization and
Society 28, 127–168.
Cooper, D. J. (1981). A Social and Organizational View of Management Accounting, in M.
Bromwich and A.G. Hopwood (eds), Essays in British Accounting Research, Pitmans,
pp. 178-205.
Cooper, D. J. , J.L. Brown, R. Greenwood and C.R. Hinings (1996). Sedimentation and Transformation
in Organizational Change: The Case of Canadian Law Firms. Organization Studies, 17: 623647.
Datar, S., Kulp, S., and Lambert, R. 2001. Balancing performance measures Journal of Accounting
Research 39 (1): 75-92.
Davis, G., Diekmann, K., & Tinsley, C. (1994). The decline and fall of the conglomerate firm in the
1980s: The deinstitutionalization of an organizational form. American sociological review, 59(4),
547-570.
Dearden, J. (1972). MIS is a mirage. Harvard business review, (1), 90-99.
Eccles, R. G. (1991). The Performance Measurement Manifesto. Harvard business review, (February),
131-138.
Ezzamel, M. (1992). Business unit & divisional performance measurement (p. 164). Academic Press.
Ezzamel, M. & Hart, H. (1987). Advanced Management Accounting (p. 352). Thomson Learning.
Ezzamel, M. and Willmott, H. 1993. Corporate governance and financial control: recent reforms in the
UK public sector. Accounting, Auditing and Accountability Journal, 6(3) 110-33.
Ezzamel, M. and Willmott, H. 1998. Accounting for team work: a critical study of group-based
systems of organizational control. Administrative Science Quarterly 43 July, 358-396.
32
Ezzamel, M., and Willmott, H. 2004a Rethinking strategy: contemporary perspectives and debates.
European Management Review, Vol. 1, No. 1, May.
Ezzamel, M., Lilley, S. and Willmott, H. (2004), “Accounting Representation and the Road to
Commercial Salvation”, Accounting, Organizations and Society, Vol. 29, pp. 783-813.
Ezzamel, M., Willmott, H. and Worthington, F. 2004. Accounting and management-labour relations:
the politics of production in the ‘factory with a problem’. Accounting, Organizations and
Society, Vol. 29, pp. 269-302.
Fiss, P. C. 2007. A set-theoretic approach to organizational configurations Academy of Management
Review, 32(4), 1180-1198.
Friedland, R., & Alford, R. R. (1991). Bringing Society Back In: Symbols, Practices, and Institutional
Contradictions. In W. W. Powell & P. J. DiMaggio (Eds.), The New Institutionalism in
Organizational Analysis (pp. 232-263). University of Chicago Press.
Hemmer, T. 1996. On the design and choice of "modern" management accounting measures. Journal
of Management Accounting Research 8 (Fall): 87-116.
Henri, J.F. 2006. Organizational culture and performance measurement systems. Accounting,
Organizations and Society 31: 77-103.
Huff, A., Jenkins, M. (Eds.), 2003. Mapping Strategic Knowledge. Sage Publications, Thousand Oaks,
CA.
Ittner, C., & Larcker, D. F. (1998). Innovations in performance measurement: trends and research
implications. Journal of Management Accounting Research, 205-238.
Ittner, C.D., Larcker, D.F., 2002. Determinants of performance measure choices in worker incentive
plans. J. Lab. Econ. 20 (2), 58.
Johnson, H.T., and R.S. Kaplan. 1987. Relevance Lost: The Rise and Fall of Management Accounting
Boston, MA: Harvard Business School Press.
Kanji, G., Moura e Sa, P., 2002. Kanji’s business scorecard. Total Quality Management. 13 (1), 13–27.
Kaplan, R., & D. P. Norton. 1992. The balanced scorecard—Measures that drive performance. Harvard
Business Review 70 (1): 71-79.
Kaplan, R.S., and D.P. Norton. 1996 The Balanced Scorecard: Translating strategy into action.
Boston, MA: Harvard Business School Press.
Kennedy, F.A. & Widener, S.A. (2008) A control framework: Insights from evidence on lean
accounting Management Accounting Research, 19, 4: 301-323
Lipe, M., Salterio, S., 2000. The balanced scorecard: judgmental effects of common and unique
performance measures. The Accounting Review 75, 283–298.
Malina, M., & Selto, F. (2004). Choice and change of measures in performance measurement models.
Management Accounting Research, 15(4), 441-469.
Malmi, T. and Granlund, M. 2009. In search of management accounting theory. European Accounting
Review 18(3): 597-620.
33
Malmi, T. and D.A. Brown. 2008. Management control systems as a package—Opportunities,
challenges and research directions. Management Accounting Research 19 (4): 287-300.
Merchant, K.A., & Van der Stede, W.A (2011). Management Control Systems. Harlow, Essex: Prentice
Hall, 2012).
Meyer, J. W. (2009). Reflections: Institutional Theory and World Society. (G. S. Drori & K. Georg,
Eds.)World Society The Writitngs of John W Meyer, (2008), 36-63. OUP Oxford.
Meyer, J. W., & Rowan, B. (1977). Meyer, John and Rowan - Institutionalized Organizations, Formal
Structure as Myth and Ceremony (1977) American Journal of Sociology, 83, 340-363.
Milgrom, P., Roberts, J., 1995. Complementarities and fit: strategy, structure, and organizational
change in manufacturing. Journal of Accounting and Economics19 (2/3), 179–208.
Miller, K. (2005). Communication theories: Perspective, processes and contexts (2nd ed). NY:
McGraw Hill.
Oakes, L., Townley, B., and Cooper, D.J. 1998. Business planning as pedagogy: language and control
in a changing institutional field Administrative Science Quarterly. June: 257-292.
Otley, D. (1980) The contingency theory of management accounting: achievement and prognosis,
Accounting, Organization and Society, 5(4), pp. 575–607.
Otley, D. 1994. Management control in contemporary organizations: Towards a wider
framework. Management Accounting Research: 289-299.
Otley, D. (1999) Performance management: a framework for management control systems research,
Management Accounting Research, 10(4), pp. 363–382.
Otley, D. and Berry, A. (1980) Control, organisations and accounting, Accounting, Organizations and
Society, 5(2), pp. 231–244.
Porter, M., 1980. Competitive strategy The Free Press, New York.
Porter, M., 1985. Competitive Advantage: Creating and Sustaining Superior Performance Free Press,
New York.
Powell, W & DiMaggio, Introduction, in Powell, W. W., & DiMaggio, P. J. (1991). The New
Institutionalism in Organizational Analysis). University of Chicago Press. doi:10.2307/258726
Rucci, A.J., Kirn, S.P., Quinn, R.T., 1998. The employee–customer–profit chain at Sears. Harvard
Business Review 76 (1), 82–97.
Sandino, T. (2007). Introducing the First Management Control Systems: Evidence from the Retail
Sector. The Accounting Review, 82(1), 265-293.
Simons, R., 1987. Accounting control systems and business strategy: an empirical analysis Accounting
Organization and Society 12, 357–374.
Simons, R., 2000. Performance Measurement&Control Systems for Implementing Strategy:
Text&Cases. Prentice-Hall, Upper Saddle River, NJ.
Solomons, D. (1985). Divisional performance: measurement and control (p. 324). Markus Wiener Pub.
34
Stern, J. M., G. B. Stewart III, and D. H. Chew, Jr. 1995. Tbe EVA® financial management system.
Journal of Applied Corporate Finance 8 (Summer): 32- 46.
Strang, D, & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and society, 22(4),
487-511.
Thornton, P. H., & Ocasio, W. (1999). Institutional Logics and the Historical Contingency of Power in
Organizations: Executive Succession in the Higher Education Publishing Industry, 1958-1990 1.
American Journal of Sociology, 105(3), 801–843.
Vancil, Richard F Decentralization, managerial ambiguity by design, Dow Jones-Irwin, Homewood
Illinois, 1979. No. of pages: 393
Figure 1: the EFQM Model
PLAN
DO
CHECK
35
Figure 2: BSC
Fig 3: Process Improvement
Mana
geme
nt
Proce
sses
36
Process Management Methods
Custo
mer
Relati
onshi
p
Mana
geme
nt
Process Management Roles and
Responsibilities
Business Processses
Suppl
y
Chain
Mana
geme
nt
Produ
ct
Lifec
ycle
Mana
geme
nt
Supp
ort
Proce
sses
Fig 4: Time: Goals-Measures-Consequences
Innovation
Initiatives
Initiative X
Customer focus
Initiatives
Initiative X
37
Global
competitiveness
Initiatives
Initiative X